Skip to content

The 10 Ways I get MASSIVE value out of Claude

Patrick Allmond · · 38 min

About this episode

What it covers

Patrick explains that the more he uses AI—primarily generative AI—the more value they get, emphasizing mindset and big-picture planning rather than using AI only for content or simple research. He mainly relies on Claude (trained for nearly a year), use ClickUp to store plans and tasks, and have recently begun testing Grok, noting AI tools are becoming commodities that copy each other. He describes AI as a fast prediction engine, not truly intelligent or creative, and share 10 practices to get better results: start with the destination, use AI as a thinking partner, provide context, let it interview you, build a blueprint first, work backwards from outcomes, have it critique your plan, use one source of truth, generate many options and curate, and turn successful prompts into systems and SOPs. He also mentions artifacts, Gantt charts, token limits, and upcoming classes/webinars.

Key moments

Jump to a topic

  1. 0:00Mindset Shift With AI
  2. 1:19Which AI Tools I Use
  3. 3:39AI Is Prediction Engine
  4. 6:26Ten Big Picture Rules
  5. 6:59Start With Destination
  6. 9:55Partner Not Typist
  7. 10:52Context Over Commands
  8. 12:25Let AI Interview You
  9. 13:22Blueprint Before Bricks
  10. 13:39ClickUp Workflow And Artifacts
  11. 16:32Work Backwards To Tasks
  12. 18:11Make AI Challenge You
  13. 21:25One Source Of Truth
  14. 22:34Generate Then Curate
  15. 27:04Systemize What Works
  16. 27:56Recap And Big Picture Wrap
  17. 33:59Community Shoutouts And Signoff
  18. 34:50Nerdy Claude Limits Story
  19. 37:30Final Thoughts And Goodbye
Transcript

Read the full episode

Captions generated by YouTube, so expect the odd typo. Select a time to play from that point.

0:00The more I use AI, and the more I learn how to use AI, and the more I change my mindset on how I use AI, which is what we're going to be really talking about today, the more value I get out of it. And um I always when I whenever I present on this topic or I teach or I do a news interview, I always like to quantify what I mean when I say I AI. Because like back in the day when we had social media, AI is this giant blanket statement. And people are thinking everything from traffic lights to those General Dynamics robots that people are kicking around to the stuff that's happening on your Mac to people using AI to generate art, to you people using gener- AI to generate those ugly ass flyers. Everyone people thinks all that stuff is people think that every single thing is AI. So I like to quantify it. So in my world, we mainly use uh my

0:52clients and I mainly use AI in terms of generative AI, which means we use AI to flush out ideas, we use AI to build applications, we we do use AI to build content, we use AI a little bit uh for um like video editing and video generation. I'm not a I'm not a really big into that world yet. Uh and we use AI a lot for from doing like big picture planning, and that's mainly what I want to talk about today. Cuz I did a lot of I've been doing and I did a lot of that recently. I basically

1:20have got to the point now where when I have an idea or a concept of something I want to build, I am spending a ton of time a ton of time talking over, flushing out, arguing with, counter doing, you know, point counterpoint, doing all that stuff with uh my main LLM, which is Claude. And and I've Claude I've been training for probably just under a year. And I know people use multiple LLMs, but I have just found that I get the most value out of focusing on one. And and I say that, but I have to admit there are a couple I use. I 100% use Claude. Claude is my main, my go-to, my boo. Claude is my boo. Um but, you know, I'm a big fan of ClickUp and the things they're building in AI. And um and so, those are probably my main two. And then I may dance a little bit on the periphery of some other ones.

2:11But yeah, Claude is my Claude is running on my computer 24/7. Uh and all I I take that back. Recently, in the past three or four days, due to the encouragement of Carrie Taff over at Taff Systems, I have looked into Grok and I am finding some really great Grok, excuse me, not Groot. I wish they would have called it Groot. That would have made it much more bearable to deal with. No, Grok. I am I am I have to admit I'm Grok is growing on me. It really is. But I'm the kind of person that you think being the high-tech history person that I have that I am with decades in in in the world of anything tech that I would be all over new stuff. No, I'm actually the opposite. I'm a slow adopter. I I wait for the fanfare to die down. I wait for bugs to be flushed out. And I wait for people to show me why I should use something different than what I'm using now. Because in the world of AI right now,

3:02everything is becoming a commodity. Which means if OpenAI can do it, Grok's going to be able to do it pretty soon, and Claude's going to be able to do it, and vice versa. And and the funny part about when I look at technology, I hearken back to the days when when social media was rolling out. And social media platforms just end up copying each other. They they really did. And I used to see that LinkedIn doing that now. And that's what that's happening in the world of AI. But I see LinkedIn adding features that Facebook has. And and if you have enough money, you can just copy everybody. And then it's just a matter of trying to you know, trying to stay trying to figure out how to stay

3:39different. So, today I want to talk to you about using big picture thinking as you work inside of AI. The coolest thing about AI is is to me is the fact that it's really simple. It's simple in terms of the fact that all AI is a prediction All AI is is a prediction engine. I have yet to see AI come up with anything new and I have yet to see anyone to show me how AI has come up with anything new that didn't already exist in the world. That's what people are afraid of. People are afraid of AI, you know, we make it to this point. I haven't seen it yet. But AI people are afraid of AI is thinking of its own unplugging itself and taking over the world. All AI is doing is taking our thought patterns and

4:29scaling them and moving faster with an AI. To me, AI I always like the phrase AI cuz it says artificial intelligence. AI is neither artificial nor is it intelligent. It's just a prediction engine. It's a really fast, really scalable prediction engine. So I when I say I want to build a program that says hello world and if that program back in the day used to take me, you know, a minute or two or or 30 minutes to build a program that says hello world, loop through 10 times and stop, I'm harkening back to my early computer days, then if I tell AI to build that program now, AI can bang it out in like 5 seconds, if that. If you know, if that. And all AI is is a prediction engine. Humans are predictable. So when you start thinking in terms of big picture in the world of of AI and the things you want to accomplish with your life and the things

5:20you want to accomplish with your business, then you realize that that what you want to do has already been done. You want to build an app? People have been building apps for years, decades, excuse me. You want to build a cool UI? People have been doing that for decades. You want to build a flyer? Heck, you want to build a car? You want to build You want to replicate something that somebody else has done? Then AI then we have number one, we have already done that as humans. Number two, we've already documented the process on how to do it. And number three, that process is online. And if it's publicly available, AI can figure out how to do it. And I kid you not when I say things like, "Can AI build a car?" Well, can AI physically build a car? No. But people have put all the steps it takes to build a car online. And you can say, "I want to build a four-person car that has a 300

6:13horsepower engine that has five tires, that gets this miles per gallon, has these features." And you can ask AI, and AI will show you how to build that freaking car. It's It's been It's happened. It's predictable at this point. So, what I want to go through

6:27here are about 10 things that that I have learned recently that help me get more value out of AI. And that's what this whole thing is about, is getting more value out of it. A lot of people are just using AI to write blog posts, write social media content, or or or doing simple research. And I really am trying to take that stuff to the next level. So, uh the first one A lot of these are going to seem like clichés, by the way. I'm going to say something and say, "Oh, Patrick, yeah, I've always heard that." Yeah, well, you know what? It's true over in the AI world. And that starts up with

6:59the first one. Number one, start with the destination, not the task. I um started uh a project recently in AI. And I'm going to try to apply all these to a project I built. And I would love your comments in the bottom if you have to go through these also. I started a new project in AI in the past week. I said, "You know what? I want to redo my family of websites. I've got, you know, this domain over here that's for this project, and this one over here that's for enterprise work, and this one over here that's for personal work. And And oftentimes, I feel like my stuff is not well connected. It's It's disjointed, cuz you know, I developed this one 10 years ago, and this one six months ago, and this one five years ago. And oftentimes, I feel that um the work I put out there is not synergistic if that's even a word that I have multiple projects going on and they

7:50don't relate to each other. So I want to kind of reimagine my universe. So I started a new conversation with AI about a month ago. Uh Claude of course over in Claude chat not Claude code. You start with Claude chat and then if it's going to turn into something code like or website like you move it over to Claude code. And I said, you know what? Here are the three areas where I I do business and I where I want to do business. And keep in mind Claude already has months and months of my mindset, my verbage, my branding, uh you know, my financial goals. Claude has all this. So when I open up Claude on the same week ago and I said, you know what? I have these three areas of my business that I want to be more synergistic and I think I might even want to rebuild their websites. And so I started with the destination

8:41not the task. That's point number one. Start with the destination not the task. Before you ask for a single I have to write this down here. Before you ask for a single deliverable define what success looks like in 6 to 12 months. Not in a week, not in two weeks, not in a month. In 6 to 12 months. Write me an email as a task. I want 15 new clients for my local market by June as a destination. AI can't aim if you don't give it a target. So that's point number one right there. Give it the end goal. I've gone into my AI AI multiple times and I've said, I want this to be my MRR or I want this to be my ARR or I want to be doing this personally in 12 months. I want to be able to, you know, do a hanging uh hanging crunch or or 50 crunches in a minute. You you give it the end goal and it works backwards. And this is something for technician

9:31it's really hard for technicians sometimes and I consider myself a technician. Is I I look at technology tools like this and I'm like, I want to tell it what to do. Step one, I want you to open up a browser. Step two, I want you to do this and this and this. And I'm having to re I'm having to re-control myself or rejig myself to think big picture in the world of AI. So, that's number one.

9:55Uh number and what this actually segues into number two here, treat AI as a thinking partner, not a typist. Most people use AI like a faster keyboard. The real value is using it like a strategist sitting across the table. Shift from do this to help me think through this. Now, I'm always very careful when I say I want AI to think for me, okay? AI doesn't think. It AI just takes and remember AI is nothing but a really good prediction engine. AI, when I tell AI to think about this or it give me ideas on something, it's just going out there and scanning millions, if not billions, of previous examples of what I want to do. And it it can scan if I say, for example, I want to build a web and application that does this, it can scan people who have built millions of applications and say and take the best of all of them and give them to me. So, that's number two. Treat AI as a

10:46thinking partner to pull in ideas from other people, not as like a secretary, if you will, okay? Uh number three, give

10:53it context, not commands. So, and this is the one I've been doing for months is AI knows what you tell it. Spend the first 10 minutes on who you are, who you serve, what you sell, your voice, your constraints, and what's already been tried. A great answer to a poorly framed question is still a bad answer. So, this is the thing that I think I I didn't realize I was doing, but now I'm realizing it, you know, the outputs I get from AI is for the longest time I've been telling AI about my voice, you know, my branding, how I think, who I want to serve, uh what I want my mission in life to be, you know, what I want to be on my tombstone, let's say, okay? I'm telling AI about these things, and then what that means is that every single thing I build takes those into consideration.

11:43You mean imagine that? Imagine you have a person sitting across from you that knows your dreams, your hopes, your goals, your fears. It knows all those things. And you could have a conversation with that person. You could say, you know what? I Now that you know all that stuff, I want to build this, or I want to sell this, or I want to develop this presentation. That person can then say, well, you know what you said 6 months ago, do you really want to do that? And I have 100% and I can hear you guys in the comments if you've had this happen, too. I have 100% had my Claude come back to me and say, you know what? You said this 6 months ago, you're saying this now, so which is it, bro? Which is it? I've had AI say that, which is it? Which one do you want to do, okay?

12:25Uh this is my next one, my fourth one is one I always like. Uh let it interview you before it builds anything. And I've I've talked about this before in previous shows, is and even the classes I gave locally here in Oklahoma City, uh tell AI to interview you. You can do that. You can tell AI, you know what? I want to build this, but I don't even know where to start. Ask me a bunch of questions, pick my brain to help me give you context. Because sometimes you may have trouble expressing yourself, and you can use these LLMs, you can use AI to pull stuff out of your brain. And that's how it learns about you. That's what I've been doing, like I said, for the past probably year in Claude, how long I've been using Claude, is I oftentimes, almost every single time I have a new idea, I'm like, ask me five to 10 questions to help me flush out this idea. So, between my big

13:15picture thinking, which it's had for months, and my five to 10 ideas for a current project, I get some pretty good output. So, let it interview you.

13:24Uh Line five, build a blueprint before the bricks. Have the AI turn your vision into a structure. Core message, audience segment, offers channel systems. That architecture becomes the map of every piece of code, content, and the marketing hangs on. What's really cool about AI,

13:41one of the coolest things that I don't see a lot of people use yet, are artifacts. And if you've ever used artifacts, let me know in the comments. When I I love it when I and I just When I say this stuff, it's like I've been using I do this stuff daily and it's happened to me in the past 24 hours. Yesterday, I was working on a marketing plan for a marketing tactic I want to execute. And what AI did is AI I gave it the timeline I want, I gave it, you know, cost, I gave it dependencies, and out of the blue, AI generated me like a Gantt chart, a colorful Gantt chart in my branding of of what the timeline would look like and what I needed to do in each phase of the of the timeline. And and now I can and now that it pushed that over into the tasks in ClickUp. Uh hey Ripley, good to see you. Um it pushed that over to the task in ClickUp.

14:32Not only did it push that image, but then I told it in ClickUp, which is my project management system, which um but no Grok yet, huh? Yeah, I just got on Grok, man. Uh but what I also told it in my project management software, and if you've watched any of this previous any of my previous shows, you know I'm a huge fan of ClickUp, is I told it build the same Gantt chart in ClickUp, add the dependencies on there, find the critical path, which is the longest, riskiest way through a project plan, where you do there. And then I was over in ClickUp and sure enough, my Gantt chart was there and my calendar, my entire project was right on my calendar there. I have Whenever I build Whenever I talk to Claude, uh whether it's Claude code or Claude chat, um it gives me an ideas and textual plans and often times I say, now go build this over in ClickUp so I can see

15:22what looks like because Claude is kind of my conversation engine where ClickUp is my repository. For example, right now my show notes I'm I have a screen right here. My show notes are in ClickUp for this show. ClickUp is where things live long-term for me right now. Claude is where things live short-term. And I have a great relationship with that company and I cannot advocate them enough. They did not sponsor anything for me, but man, I would I would love love for you guys to to give it a shot. And if you have a And you know what? I like to tell people, you know what? If you have Asana or Monday or something else, you know what? Feel free to use it, but use the heck out of it, okay? ClickUp, I've tried every single thing I just mentioned and ClickUp runs circles around them. Even even the stuff coming out in HighLevel. HighLevel is coming out with a project management system. I am I am fairly certain they're basing a lot

16:12of what I've seen on ClickUp because I know HighLevel uses ClickUp internally. And so so I will I will stick with I don't really care what HighLevel comes out with the project management company. This is one of those areas where a dedicated tool will run circles around something else. So I'm going to stick with my ClickUp. So anyways, um uh so we talked about uh building a blueprint from the bricks. Number six,

16:33work backwards from the outcome. Reverse engineer it. Take from your destination to your quarterly milestones to your monthly projects to this week actions. Ask your AI, if this goal is achieved, what had to be tuned 90 days before, 30 days before this week. This is where big ideas come into a to-do list. And again, I would take all that stuff and I would reverse it into something that can be in a list. And whether you maintain your tasks in HighLevel or in Monday or in Teamwork or Asana, it doesn't matter. I do mine in ClickUp, but oftentimes when I when I tell AI to do something like this, work backwards, I say lay out all the tasks and the deadlines and assign them to me. Sometimes I meet those deadlines, sometimes I don't. But but it's really you I don't just like I don't like using AI to just give me a list of plans that I copy paste over

17:24into a doc and then disappears. No, if you've ever worked with me, you know, ClickUp is my source of truth when it comes to to plans and I will tell Claude this cool plan you've built for me, this cool 10 20 50 6 step plan, go dump it into ClickUp, assign it to me, put all of your comments in the comments there. Because you to me um often times I will I will I will go down the rabbit hole in a Claude chat and I I I may sit in the Claude chat for days and come out with all kinds of things, but then again I start other Claude chats and those ones get pushed down. So now my list of Claude pin chats is getting quite long. I really like to have Claude push that over into ClickUp for ClickUp being my source of my source of truth. Uh so number six was work backwards from the outcome. Number seven, make AI argue

18:13with you. And this is something I've seen a lot of people have a lot of fun with is AI by default is it doesn't have feelings, but the way they programmed it it tends to want to please you, if you can use that emotion in the world of AI. AI tends to want to please you. So AI agrees with you by default, so tell it run a pre-mortem. It's a year from now and this failed, why? Or poke holes in this plan like a skeptical investor. Ideas that survive the attacks are worth building, okay? And I'm kind of curious in the comments, have you ever asked AI to argue with you or asked AI to poke holes in your plan? I don't do that as much as I should, but what I've learned is the more information I give it the the better guidance I give it, I mean the better guidance I get out of it. So I don't

19:03basically make it argue with you so much, I make it quiz me a lot. Occasionally there is kind of a I I have it poke holes in a plan, but I tend to I tend to give it so much input and so much realistic stuff like um uh like like often times I'll give it I have to give it timelines of stuff when I know things are or are not going to happen. Um uh and for example, I have a cruise coming up and so I my uh Claude right now knows that I'm going to be out of pocket. So any plan any timelines that um involve the last week of the month, it knows to push those timelines out. Not only that, be very realistic. And this one isn't even in my list. I learned this from project management training. Be very realistic with your time, okay? You do not have 8 hours of productive time every month. It's unlikely that you

19:53have 6 hours of productive time Excuse me, every [clears throat] every day. It's unlikely that you have 6 hours. You'd be lucky to have 3 to 4 hours of productive time every single day. So when you were sketching out plans or grand visions, be realistic about that. That's one of the things I learned in project management training is that the duration of something is different than the effort of something. Something may take a week even if the effort's only 5 or 6 hours. Because you have a hundred of things going on and and an hour-long what what it can what normally seems like an hour-long task may take you 2 or 3 hours sometimes. Be very realistic about your time management. That that's I it All my calendar applications, I'm very realistic. You know what? I can't

20:44take me for example, um uh my calendar software that allows people to book my time. I do not want to be able to take appointments from the moment I wake up to the moment I go to sleep. And over time this has actually shrunk. I wake up at 5:00 in the morning and I go to bed at 9:00, okay? So realistically, you know what? You should set your calendar up for that way. Nope. 8:00 to 5:00? Nope. I think I have my calendar set up so you can't book before 9:00 and you can't book after 3:00. Be realistic about your time management. The more genuine factual information uh you you give uh AI, the better the plans will come out of it. Okay, let's keep on going on. Uh number seven was make AI argue with you. Number eight,

21:26create one source of truth and let everything flow from it. Turn the strategy into a single brief or document. Every prompt for code, a landing page, a video script, or an ad references it. That's how you get consistency across 50 assets instead of 50 disconnected ones. And this is where I have a problem with people that that are using, you know, five or 10 different kinds of AI tools is that they each one has a different version of you. And there's no real way to keep them in sync that I've seen. Uh I've tried Obsidian. Obsidian and it doesn't it doesn't really do okay to me because I don't I don't I use Obsidian to map things, but I don't use Obsidian as like a data source of truth. Um that's one of the reasons I really love sticking with Claude is uh versus switching to something else is because

22:17it's like I said, I've been chatting with it for a year. It knows my hopes, my dreams, my failures, my weaknesses, my strengths, and everything gets filtered through that. Um number nine as we here as we slowly come towards the end here. Okay, so number eight was create one source of truth and let everything flow from it.

22:34Number nine, generate in volume, curate with judgment. So, what this means is that it's okay to throw a lot of spaghetti against the wall. AI is cheap at making options. Your taste is the scarce resource. Ask for 20 headlines, five approaches, three versions. Your job shifts from creator to editor-in-chief. You're the one who knows what's right. Man, I I love all these cuz I've done all these. And comment below if you've done that. Comment below if you've done something like give me 10 ideas for an ad or give me 20 ideas for a an ad headline. I'll give you an example here. I oftentimes use Claude for copy for clients. And you know on the Facebook ads you have what you have the headline, you have the description, and you have something else. Whenever you build a Facebook ad there's three

23:26there's three possible sections where you can put in text. There's like a short one, a longer one, and a a long one, really long one which is a description. And you can fill in all three. Not only can you fill in all three, you can do multivariate testing, which means I can give Facebook five headlines and it will run through five headlines as it generates different iterations of the ads. So if you think about that, think about that you get you you get five headlines, you get five descriptions, and you get five whatever the other one is, okay? So you have what? 5 * 5 * 5, 25 * 5, you have 125 possible combinations of people things people see in your ad along with whatever video and for the love of all that's holy, please make sure you're doing video in your Facebook ads. So what I will often do is after I've had a

24:16conversation with Claude and I have flushed out kind of how what the goal of the ad is of the ads are, I will go to Claude and I will say give me 15 each of these. Give me 15 headline options, give me 15 descriptions, and give me 15 of the other thing, whatever it is, I can't think of it right now. And it will do that. It cuz remember I've been talking with Claude for days. It knows what I want to sell. Okay? It knows all that. So what I do is I say I want 15 variations of these things in my ad copy or let's say it's a headline for a real or headline for a YouTube video like this. Um I will I will have Claude give me 15 variations of that and from those 15 15 15 I will pick five of the headlines five of the descriptions and five of the other things and I will say these are

25:06the five which to me personally seem to resonate well and I'll go over to my Facebook ads and I will put those in there okay. I'll put the five headlines I'll put the five the ones that I filtered out uh the five headlines the five descriptions and the five of the other things over in my Facebook ads and now Facebook can then churn through those and figure out which one performs best. And what's cool about when you when you do those 15 over in Claude is is at the bottom summary Claude will tell you I recommend you know number one six and five on descriptions or I recommend number two three and four on on headlines or on topics or headlines. I think it's the headlines one of them too. Um and so Claude will even give you the recommendation. Sometimes I go with the Claude one because it has maybe it gives me a perspective I hadn't seen. Like it'll it'll tell you oh number three has

25:57more of an emotional one but number two is more of a kind of a call to action one. So so when it comes to coming up with ideas for copy the these are areas where Claude gives you so many options and you can churn through ideas so much faster than other people can but you have to generate in volume. If you have five if you're over in a platform here and you want Claude to generate something for you to put over this platform ask it for multiple variations. Ask it for multiple variations for different themes. Ask it for you know you know five uplifting variations five mediocre variations and and five variations uh twisting someone's pain. Often times when I'm building you know ads uh I learned a while back that people buy for two reasons. People buy because they want to seek pleasure or

26:47they want to avoid pain. I will often use that when I'm generating ad copy to get specific iterations. Okay. Uh so again, that was number nine, generate in volume, curate with judgment. And number 10 here, and again, I'm kind of curious in the comments, let me know if you've used any of these. Number 10 here, turn what

27:06works into systems. When a prompt or workflow or format works, save it as a template. A custom instruction or a skill. If you haven't done so, please make sure you start a prompt library. That's me talking to you right now, not in my notes, okay? Start a prompt library. I even have a prompt library on my that I save that's actually public on my website. And if you want that, you know, DM me or comment some place and I'll get you that. But there's a link on my website that I I have found a bunch of prompts. I save them in a ClickUp doc, and then I share that ClickUp doc on the web if you guys want that. So when a prompt, workflow, or format works, save it as a template. A custom instruction or a skill. Then the next project starts at step eight, not step one. And this is where AI goes from a novelty to a compounding advantage, okay?

27:56So quick recap on these as you go into your amazing Wednesday. Uh start with the destination, not the task. Treat AI as a thinking partner, not a typist. Give it context, not commands. Let it interview you before it builds anything. Build the blueprint before the bricks. Work backwards from the outcome. Make AI argue with you. That's always fun. Create one source of truth and let everything flow from it. Generate in volume, curate with judgment, and finally, turn what works into systems and into SOPs. And again, that's an area where I just love the fact that I use Claude with ClickUp because if I if I generate a really cool SOP in Claude, I can then say, "Oh, now go over this and put this in the SOP

28:47section inside of my ClickUp." And bam, the team has it as an SOP. Okay? The people that are doing really amazing things with AI um aren't aren't the ones that that are doing that um are coming up with better prompts. They're They're the people who are good at giving AI big picture ideas. And telling it where they want to be. Telling it where they're going. Telling it their resources they have available. You can do stuff like that. You can in AI, you can say, "This is where I want to be in 6 months." By default, AI will tell you, "Oh, I can do all these grandiose plans for you." But then you have to factor in time.

29:37You have to factor in your budget. You have to factor in resources. And the more information you give it, the better the realistic plan is that comes out of these discussions. And again, I'm I'm I'm Hopefully, I'm preaching to the choir. And if you've done this, please let me know in the comments or with a like or give me some kind of sign you've done this. I have When I say I do this on a regular basis, I am not kidding. In the past 24 hours, I made up a marketing plan um that I I basically Where did I Where did I make that up in? Did I do it in Claude? I think I did someplace in a combination between Claude and Grok because I have been testing Grok. And then I but I connect all my stuff to ClickUp. And I say, "Here's my constraints. Here's my time constraints." For example, you know, I think I've told everyone, you know, multiple times I'm going on a

30:27cruise the last week of this month. That is a huge time constraint. I don't want to be having to worry about doing anything on that cruise. So, that's an unavailable pocket of time for me. You know, weekends are unavailable pockets of time, vacations, birthdays, things like that. Give it constraints as far as time. Give it constraints as far as resources. Give it constraints as far as revenue or, you know, marketing budget. Feed all that into your single source of truth, and then what was which the one was a single source of truth, right? That was Oh, that was number eight. Create one source of truth, then let everything flow from it. And number eight here, the create one source of truth, that's why I'm such That's why I'm a guy who's very apprehensive about throwing new AI tools in my mix because every single time I throw on a new AI tool, I either have to export it from one export memory from one or, you know, a branding document

31:17from one and feed it into the other. And the two often do not agree. So, I'm going to I'd be kind of curious if anyone's ever done that. If anyone has who's using multiple AI systems and they how they keep the two in sync. That to me would be would be very a very fascinating uh a very fascinating thing to to talk to someone about. And again, remember the people winning and how I want you to win with AI is give it big picture stuff and back into the the strategies and the tactics. As a matter of fact, oftentimes you hear start with strategies and then work over in the tactics. I actually like doing something one thing bigger. This is what I'll wrap up with here. And this is actually going to be part of a website design which, ironically, I am building in in uh in uh in Claude. Okay, I went to Claude and I said, "Here's how I want my website to be and here's how I want it to represent

32:07my ideas." And so, one of the designs that I wish I could show you here in Claude, one of the images that's going to be on my homepage is the one icon for your big idea broken down into multiple strategies and then broken down into multiple AI agents. Oftentimes you hear start with the strategy and then break it down into the tactics. I actually want to start really big picture, start with the big giant goal, start with your grand idea, then strategies, and then break it down into tactics. So, if you uh if you've done any of this or you have some other ideas as long as I didn't uh cover, uh please uh drop me a comment. If you don't feel like leaving your name in the comment, DM me and I'll talk about it on a future show, but I want you to succeed with AI. And if you have any questions about how to get better with AI, uh you know, feel free to DM me. You know, as long as I

32:57have the time, as long as you don't expect an immediate answer, I will gladly, you know, send you a message of some ideas on how you can better use these tools. Um and if you're in Oklahoma City, definitely ping me because we do run uh in-person classes on occasion, and I've been told I need to start doing more webinars. So, that's going to be in the future. And as always, if you have not done so, hop over to YouTube and subscribe, like, follow, share, do all the cool things that allow me to spend my time making you a better person and making me a better person. Cuz I learn every single time I learn from my lessons where I teach you. That's one of the reasons that I love teaching is cuz I learn something every single time I teach something. And I swear if if, you know, if if I could I I I that I don't know I don't know how to put it any any better

33:47than that. I love teaching because I learn when I teach. I have to learn whatever I whatever I talk about so I can make you a better person, but I become a better person 5 minutes before you become a better person. Mr. Kennedy,

34:00I see you there. Man, I don't know. Hopefully I'm not interrupting your live cuz I know you do a live pretty regular also. But Mr. Kennedy, Robert Kennedy the III, ladies and gentlemen, if you have not done so, hop over and make sure you follow this guy, too. He's one of the people I absolutely love. My brother from another mother, heart and soul, Mr. Kennedy. He has a live show also on YouTube and on Facebook, and you definitely pay attention to the things he's putting out there because um he is uh he's he's a pretty sharp guy. Richard Richard, sir, good to see you. Good to see you, man. Oh, goodness gracious. See? And here I am fumbling with the account. Robert, you get that, man. Anyways, uh everyone have a blessed day. I'm going to polish off my coffee. I'm going to go back. I've got a long-running project in Claude. I've got a long-running project in Claude. For the first time, okay, the

34:52formal part of the show is wrapped up. I'm going to share something nerdy with you guys, okay? Let me know if you're hanging around here. Uh let me know if you're hanging around watching this. For the first time ever in Claude, I am getting warnings that I'm getting ready to hit my limits. For the first time ever. In the past two or three days in Claude, I have chewed up more usage and more tokens than probably the past two or three months. I have never had a Claude usage warning before. I'm hitting them now. And that's because I gave it uh Richard, dude, good to see you. Um that's because over the past week, I've been giving it some big picture idea stuff and having it generate, um you know, um assets out of that. Have it generate artifacts out of that. And also for the last time yesterday, or for the first

35:42time ever yesterday, I gave um Hey Rick, man, good to see you. Haven't seen that name in a while. I'm back watching. Hey Rick, please. Um um for the first time ever, I've been And this goes back to what I was talking about earlier. I I keep been giving Claude big picture ideas. And when you do that, and you then say, "Okay, go off and start start creating start implementing my big picture idea." You'll notice that Claude just disappears for hours at a time. It gives you updates, but it's churning on ideas. And that happened yesterday. I kid you not. I I um you know, I have I have I have multiple monitors here and I was I was working in Claude, and Claude was working on something for like an hour, And I kept going back over the window like is Claude asking me for something? Nope. Claude's I'm building this. I'm

36:33working on this for you. I'm building this. And I even prompted him. I'm like, are you building anything for me? And Claude's like, nope. I got you, fam. I got you. I'm working on this for you right here. And sure enough, I So, I decided to go for a walk yesterday evening. I went for a walk for uh I don't know. It's on my phone. It was at least 35 minutes. I went out and did walk, did some push-ups. I came back. Claude was still working. And it wasn't stuck. It wasn't stuck. And to me, that's just fascinating. It really is. I Again, if you guys know, I'm not a neophyte when it comes to technology. I've been in this game for a long time. And every single day I am just amazed and fascinated at the tools that are at our fingertips. I really am. I mean, look at this. I've got lights and cameras and microphones, and I can talk to anybody in the world from here. I've got a Claude agent in the background that's building software for me and generating ideas

37:25for me to help make the world a better place and have more of an impact. We live in an amazing time.

37:31But let's keep it amazing. Go out and vote on November 2nd. Have some coffee today. Make sure you stretch and hydrate today. And I will see you next week.

More videos
5:01

Video · November 10, 2026

Claude Duets: My First Look

Claude Duets: My First Look: watch the episode and read the full transcript.

Watch · 5 min

6:30

Video · November 5, 2026

Why I'm Using the New HighLevel Workflow Map

Workflow Map Demo: Visualize Automations, Tags, Pipelines & Use Ask AI to Fix Issues The script demos a new Workflow Map view in a test account, showing how…

Watch · 7 min

See all videos →

Want this working in your business?

Book a call with a marketing specialist, or just text us if you have questions.